Development of Attachment

 Attachment is the strong, affectionate tie we have with special people in our
lives that leads us to feel pleasure when we interact with them and to be
comforted by their nearness during times of stress.
 By the second half of the first year, infants have become attached to familiar
people who have responded to their needs. Watch how babies of this age
single out their parents for special attention. When the mother enters the
room, the baby breaks into a broad, friendly smile. When she picks him up,
he pats her face, explores her hair, and snuggles against her body. When he
feels anxious, he crawls into her lap and clings closely.
 Freud first suggested that the infant's emotional tie to the mother provides
the foundation for all later relationships. Although Freud was correct that the
quality of the infant—parent bond is vitally important, contemporary
research indicates that his ideas merit revision: The contribution of
attachment to long-term development depends not just on the infant's early
experiences but also on the ongoing parent—child relationship.
 Psychoanalytic perspective regards feeding as the central context in which
caregivers and babies build this close emotional bond. Behaviorism, too,
emphasizes the importance of feeding, but for different reasons. According
to a well-known behaviorist explanation, as the caregiver satisfies the baby's
hunger (primary drive), the infant learns to prefer her soft caresses, warm
smiles, and tender words of comfort (secondary drive) because these events
have been paired with tension relief.
 Although feeding is an important context in which mothers and babies
build a close relationship, attachment does not depend on hunger
satisfaction.
 In the 1950s, a famous experiment showed that rhesus monkeys reared
with terry-cloth and wire-mesh "surrogate mothers" clung to the soft
terry-cloth substitute, even though the wire-mesh "mother" held the
bottle and infants had to climb on it to be fed (Harlow & Zimmerman,
1959).
https://www.youtube.com/watch?v=_O60TYAIgC4&feature=emb_logo

 Similarly, human infants become attached to family feed them,

including fathers, siblings, and grandparents. And you may have noticed
that toddlers in Western cultures who sleep alone and experience
frequent daytime separations from their parents sometimes develop
 strong emotional ties to cuddly objects, such as blankets or teddy bears.
Yet such objects have never played a role in infant feeding!

Bowlby's Ethological Theory

 Today, ethological theory of attachment, which recognizes the infant's

emotional tie to the caregiver as an evolved response that promotes survival,
is the most widely accepted view.

 John Bowlby (1969), who first applied this idea to the infant—caregiver
bond, was originally a psychoanalyst. In his theory, he retained the
psychoanalytic idea that quality of attachment to the caregiver has profound
implications for the child's feelings of security and capacity to form trusting
relationships.

 At the same time, Bowlby was inspired by Konrad Lorenz's studies of

imprinting (Lorenz 1935) investigated the mechanisms of imprinting, where
some species of animals form an attachment to the first large moving object
that they meet. This process suggests that attachment is innate and
programmed genetically. https://www.youtube.com/watch?v=ihh1xBXwt_0

 He believed that the human infant, like the young of other animal species, is
endowed with a set of built-in behaviors that keep the parent nearby to
protect the infant from danger and to provide support for exploring and
mastering the environment (Waters & Cummings, 2000).

 Contact with the parent also ensures that the baby will be fed, but Bowlby
pointed out that feeding is not the basis for attachment. Rather, the
attachment bond has strong biological roots. It can best be understood in an
evolutionary context in which survival of the species—through guaranteeing
both safety and competence—is of utmost importance.

 According to Bowlby, the infant's relationship with the parent begins as a set
of innate signals that calls the adult to the baby's side. Over time, a true
affectionate bond develops, supported by new emotional and cognitive
capacities as well as by a history of warm, sensitive care. Attachment
develops in four phases:

I. Pre-attachment phase (birth to 6 weeks). Built-in signals—grasping,

smiling, crying, and gazing into the adult's eyes—help bring newborn
babies into close contact with other humans. Once an adult responds,
infants encourage her to remain nearby because closeness comforts
 them. Babies of this age recognize their own mother's smell and voice,
and they will soon recognize her face. However, they are not yet
attached to her, since they do not mind being left with an unfamiliar
adult.
II. "Attachment-in-the-making" phase (6 weeks to 6—8 months).
During this phase, infants respond differently to a familiar caregiver
than to a stranger. For example, the baby smiles, laughs, and babbles
more freely with the mother and quiets more quickly when she picks
him up. As infants interact with the parent and experience relief from
distress, they learn that their own actions affect the behavior of those
around them. Babies now begin to develop a sense of trust—the
expectation that the caregiver will respond when signaled— but they
still do not protest when separated from her.
III. "Clear-cut" attachment phase (6—8 months to 18 months—2 years).
Now attachment to the familiar caregiver is evident. Babies display
separation anxiety—they become upset when the adult on whom
they have come to rely leaves. Separation anxiety does not always
occur; like stranger anxiety, it depends on infant temperament and on
the current situation. But in many cultures, it increases between 6 and
15 months. Its appearance suggests that infants have a clear
understanding that the caregiver continues to exist when not in view.
Consistent with this idea, babies who have not yet mastered Piagetian
object permanence usually do not become anxious when separated
from their mothers (Lester et al., 1974). Besides protesting the parent's
departure, older infants and toddlers try hard to maintain her presence.
They approach, follow, and climb on her in preference to others.
IV. Formation of a reciprocal relationship (18 months—2 years and on).
By the end of the second year, rapid growth in representation and
language permits toddlers to understand some of the factors that
influence the parent's corning and going and to predict her return. As a
result, separation protest declines. Now children start to negotiate with
the caregiver, using requests and persuasion to alter her goals, For
example, one 2-year-old asked her parents to read a story before
leaving her with a baby-sitter. The extra time with her parents, along
with a better understanding of where they were going ("to have dinner
with Uncle Charlie") and when they would be back ("right after you
go to sleep"), helped this child withstand her parents' absence.

 With age, children depend less on the physical proximity of caregivers and
more on a sense of confidence that they will be accessible and responsive in
times of need.
 According to Bowlby (1980), out of their experiences during these four
phases, children construct an enduring affectionate tie that they use as a
secure base in the parents' absence. This image serves as an internal
 working model, or set of expectations about the availability of attachment
figures, their likelihood of providing support during times of stress, and the
Self's interaction with those figures. The internal working model becomes a
vital part of personality, serving as a guide for all future close relationships
(Bretherton & Munholland, 1999).
 With age, children constantly “update” or revise and expand—the internal
working model as their cognitive, emotional, and social capacities increase,
as they interact further with parents, and as they form other close bonds with
adults, siblings, and friends.

Measuring the Security of Attachment

 Although virtually all family-reared babies become attached to a familiar
caregiver by the second year, the quality of this relationship differs from
child to child. Some infants appear secure in the presence of the caregiver;
they know they can count on her for protection and support. Others seem
anxious and uncertain.
 A widely used laboratory technique for measuring the quality of attachment
between 1 and 2 years of age is the Strange Situation. In designing it, Mary
Ainsworth and her colleagues (1978) reasoned that if the development of
attachment has gone well, infants and toddlers should use the parent as a
secure base from which to explore an unfamiliar playroom. In addition,
when the parent leaves, an unfamiliar adult should be less comforting than
the parent. The Strange Situation takes the baby through eight short episodes
in which brief separations from and reunions with the parent occur (see
Table 10.2). https://www.youtube.com/watch?v=QTsewNrHUHU

 Observing the responses of infants to these episodes, researchers have

identified a secure attachment pattern and three patterns of insecurity; a few
 babies cannot be classified (Ainsworth et al., 1978; Barnett & Vondra, 1999;
Main & Solomon, 1990). Although separation anxiety varies among the
groups, the baby's reunion responses define attachment quality.
i. Secure attachment. These infants use the parent as a secure base.
When separated, they may or may not cry, but if they do, it is because
the parent is absent and they prefer her to the stranger. When the
parent returns, they actively seek contact, and their crying is reduced
immediately. About 65 percent of North American infants show this
pattern.
ii. Avoidant attachment. These infants seem unresponsive to the parent
when she is present. When she leaves, they usually are not, distressed,
and they react to the stranger in much the same way as to the parent.
During reunion, they avoid or are slow to greet the parent, and when
picked up, they often fail to cling. About 20 percent of North
American infants show this pattern.
iii. Resistant attachment. Before separation, these infants seek closeness
to the parent and often fail to explore. When she leaves, they are
usually distressed, and on her return, they mix clinginess with angry,
resistive behavior, struggling when held and sometimes hitting and
pushing. In addition, many continue to cry and cling after being
picked up and cannot be comforted easily. About 10 to 15 percent of
North American infants show this pattern.
iv. Disorganized/disoriented attachment. This pattern reflects the
greatest insecurity. At reunion, these infants show confused,
contradictory behaviors. They might look away while being held by
the parent or with flat, depressed emotion. Most communicate their
disorientation with a dazed facial expression. A few cry out after
having calmed down or display odd, frozen postures. About 5 to 10
percent of North American infants showed this pattern.

Infants' reactions in the Strange Situation closely resemble their use of

the parent as a secure base and their response to separation and
reunion at home (Blanchard & Main, 1979; Pederson & Moran,
1996). For this reason, the procedure is a powerful tool for assessing
attachment security.

The Attachment Q-Sort is an alternative method, suitable for children between 1

and 4 years of age, that permits attachment to be assessed through observations in
the home (Waters et al., 1995). An observer (the parent or a highly trained
informant) sorts a set of 90 descriptors of child behavior—such as "Child greets
mother with a big smile when she enters the room," "If mother moves very far,
child follows along," and "Child uses mother's facial expressions as a good source
of information"—into nine categories, ranging from highly descriptive to not at all
descriptive of the child. Then a score is computed that indicates where the child
falls along a continuum ranging from high to low security. Because the Q-Sort taps
a wider array of attachment-related behaviors than the Strange Situation, it may
better reflect the parent—infant relationship in everyday life.
The Q-sort method is time consuming, requiring a nonparent informant to spend
several hours observing the child before sorting the descriptors, and it does not
differentiate between types of insecurity. But the Q-Sort responses of expert
observers correspond well with infants' secure-base behavior in the Strange
Situation. Parents' Q-Sort judgments, however, show little relationship with
Strange Situation assessments (van IJzendoorn et al., 2004). Parents of insecure
children, especially, may lack skills for accurately reporting their child's
attachment behaviors.
Cultural Variations
 Cross-cultural evidence indicates that attachment patterns may have to be
interpreted differently in certain cultures.
 For example, a study reveals, German infants show considerably more
avoidant attachment than American babies do. But German parents
encourage their infants to be non-clingy and independent, so the babies’
behavior may be an intended outcome of cultural beliefs and practices
(Grossmann et al., 1985).
 In contrast, a study of infants of the Dogon people of Mali, Africa,
revealed that none showed avoidant attachment to their mothers (True,
Pisani, & Oumar, 2001). Even when grandmothers are primary caregivers
(as is the case with first-born sons), Dogon mothers remain available to
their babies, holding them close and nursing them promptly in response to
hunger and distress.
 Japanese infants, as well, rarely show avoidant attachment. An unusually
high number are resistantly attached, but this reaction may not represent
true insecurity. Japanese mothers rarely leave their babies in others' care,
so the Strange Situation probably creates greater stress for them than for
infants who frequently experience maternal separations (Takahashi, 1990).
Also, Japanese parents value the infant clinginess and attention seeking
that are part of resistant attachment, considering them to be normal
indicators of infant closeness and dependency (Rothbaum et al., 2000a).
 Despite these cultural variations and others, the secure pattern is still the
most common attachment classification in all societies studied to date (van
IJzendoorn & Sagi, 1999).
Factors Affecting Attachment Security
What factors might influence attachment security? Researchers have looked
closely at four important influences:
1. Opportunity to establish a close relationship;
2. quality of caregiving;
3. the baby's characteristics; and
4. Family context, including parents' internal working models.

OPPORTUNITY FOR ATTACHMENT

 What happens when a baby does not have the opportunity to establish a
close tie to a caregiver?
 In a series of studies, René Spitz (1946) observed institutionalized infants
who had been given up by their mothers between 3 and 12 months of age.
The babies were placed in a large ward where each shared a nurse with at
least seven other babies. In contrast to the happy, out- going behavior they
had shown before separation, they wept, withdrew from their surroundings,
lost weight, and had difficulty sleeping. If a consistent caregiver did not
replace the mother, the depression deepened rapidly.
 These institutionalized infants experienced emotional difficulties because
they were prevented from forming a bond with one or a few adults (Rutter,
1996).
 In another study, which supports this conclusion, researchers followed the
development of infants in an institution that offered a good caregiver—child
ratio and a rich selection of books and toys. Nevertheless, staff turnover was
so rapid that the average child had a total of 50 caregivers by the age of 4 1/2!
Many of these children became "late adoptees" who were placed in homes
after age 4. Most developed deep ties with their adoptive parents, indicating
that a first attachment bond can develop as late as 4 to 6 years of age (Tizard
& Rees, 1975).
 But these children were also more likely to display attachment difficulties,
including an excessive desire for adult attention, "overfriendliness" to
unfamiliar adults and peers, failure to check back with the parent in new
situations, and few friendships. Adopted children who spent their first 6
months or more in deprived Romanian institutions frequently have similar
relationship problems (Hodges & Tizard, 1989; O'Connor et al., 2003).
Although follow-ups into adulthood are necessary to be sure, these results
leave open the possibility that fully normal development depends on
establishing close ties with caregivers early in life.
QUALITY OF CAREGIVING
 Dozens of studies report that sensitive caregiving—responding promptly,
consistently, and appropriately to infants and holding them tenderly and
carefully—is moderately related to attachment security in both biological
and adoptive mother—infant pairs and in diverse cultures and SES groups
(DeWolff & van IJzendoorn, 1997; Posada et al., 2002, 2004; Stams, Juffer,
& van IJzendoorn, 2002; van IJzendoorn et al., 2004).
 In contrast, insecurely attached infants tend to have mothers who engage in
less physical contact, handle them awkwardly, behave in a "routine"
manner, and are sometimes negative, resentful, and rejecting (Ainsworth et
al., 1978; Isabella, 1993; Pederson & Moran, 1996).
 Also, in several studies of North American babies, a special form of
communication called interactional synchrony separated the experiences of
secure from insecure babies. It is best described as a sensitively tuned
"emotional dance," in which the caregiver responds to infant signals in a
well-timed, rhythmic, appropriate fashion. In addition, both partners match
emotional states, especially the positive ones (Isabella & Belsky, 1991;
Kochanska, 1998). Earlier we saw that sensitive face-to-face play, in which
interactional synchrony occurs, helps infants regulate emotion. But
moderate adult—infant coordination is a better predictor of attachment
security than "tight" coordination, in which the adult responds to most infant
cues (Jaffe et al., 200). Perhaps warm, sensitive caregivers use a relaxed,
flexible style of communication in which they comfortably accept and repair
emotional mismatches, returning to a synchronous state.
https://modules.ilabs.uw.edu/module/attachment-in-practice/interactional-
synchrony/

 In addition, the way different cultures view maternal sensitivity depends on

their values and goals for children's development. Among the Gusii people
 of Kenya, mothers rarely cuddle, hug, or interact playfully with their babies,
although they are very responsive to their infants' needs. Yet most Gusii
infants appear securely attached, using the mother as a secure base (Leffi_ne
et al., 1994). This suggests that security depends on attentive caregiving, not
necessarily on moment by-moment contingent interaction.
 Compared with securely attached infants, avoidant babies tend to receive
over stimulating care. Their mothers might, for example, talk energetically
to them while they are looking away or falling asleep. By avoiding the
mother, these infants try to escape from overwhelming interaction.
 Resistant infants often experience inconsistent care: Their mothers are
unresponsive to infant signals, but when the baby begins to explore, they
interfere, shifting the infant's attention back to themselves. As a result, the
baby is overly dependent as well as angry at the mother's lack of
involvement (Cassidy & Berlin, 1994; Isabella & Belsky, 1991).
 When caregiving is highly inadequate, it is a powerful predictor of
disruptions in attachment. Child abuse and neglect are associated with all
three forms of attachment insecurity. Among maltreated infants,
disorganized/disoriented attachment is especially high (Barnett, Ganiban, &
Cicchetti, 1999; van IJzendoorn, Schuengel, & Bakermans; Kranenburg,
1999).
 Depressed mothers and parents suffering from a traumatic event, such as
loss of a loved one, also tend to promote the uncertain behaviors of this
pattern (Teti et al., 1995; van IJzendoorn, 1995). Observations reveal that
some of these mothers display frightening, contradictory, and unpleasant
behaviors such as looking scared, mocking or teasing the baby, holding the
baby stiffly at a distance, roughly pulling the baby by the arm, or seeking
reassurance from the upset child (Goldberg et al., 2003; Lyons-Ruth,
Bronfman, & Parsons, 1999; Schuengel, Baker-mans-Kranenburs & van
IJzendoorn, 1999). The baby's disorganized behavior seems to reflect a
conflicted reaction to the parent, who sometimes comforts but at other times
arouses fear.

INFANT CHARACTERISTICS
 Because attachment is the result of a relationship that builds between two
partners, infant characteristics should affect how easily it is established.
 We saw that pre-maturity, birth complications and newborn illness make
caregiving more taxing.
 In stressed, poverty-stricken families, these difficulties are linked to
attachment insecurity (Wille, 1991).
 But when parents have the time and patience to care for a baby with special
needs and view their infants positively, at-risk newborns fare quite well in
attachment security (Cox, Hopkins & Hans, 2000; Pederson & Mofan,
1995).
 Infants also vary considerably in temperament, but its role in attachment
security has been intensely debated. Some researchers believe that infants
who are irritable and fearful may simply react to brief separations with
intense anxiety, regardless of the parent's sensitivity to the baby (Kagan,
1998). Consistent with this view, emotionally reactive, difficult babies are
more likely to develop later insecure attachments (van IJzendoorn et al.,
2004; Vaughn & Bost, 1999).
 But other evidence suggests that caregiving is involved in the relationship
between infant difficultness and attachment insecurity.
 In a study extending from birth to age 2, difficult infants more often had
highly anxious mothers, a combination that often resulted in a
"disharmonious relationship" in the second year—characterized by both
maternal insensitivity and attachment insecurity.
 Infant difficultness and maternal anxiety seemed to perpetuate one another,
impairing caregiving and the security of the parent—infant bond.
 In another investigation that focused on disorganized/disoriented babies,
negative emotional reactivity increased sharply between 12 and 18 months.
Attachment disorganization was not caused by a difficult temperament;
instead, it preceded difficultness and seemed to promote it (Barnett,
Ganiban, & Cicchetti, 1999).
 Indeed, extensive research confirms that caregiving can override the impact
of infant characteristics on attachment security.
 When researchers combined data from more than a thousand mother—infant
pairs, they found that maternal problems—such as mental illness, teenage
parenthood, and child abuse—were associated with a sharp rise in
attachment insecurity. In contrast, child problems—ranging from
prematurity and developmental delays to serious physical disabilities and
psychological disorders—had little impact on attachment quality (van
IJzendoorn et al., 1992).
 Finally, if children's traits determined security of attachment, we would
expect attachment to be at least moderately heritable, just as temperament is.
Yet twin comparisons reveal that the heritability of attachment is virtually
 nil (O'Connor & Croft, 2001). Identical twins are not more alike than
fraternal twins (or other siblings) in attachment security. About two-thirds of
siblings—whether identical twins, fraternal twins, non-twin siblings,
unrelated siblings, or foster infants—establish similar attachment patterns
with their parent. Yet these siblings often differ in temperament (Dozier et
al., 2001; van IJzendoorn, 1995). This suggests that the strongest
environmental influences on attachment security are non-shared experiences,
reflecting most parents' efforts to adjust their caregiving to each child's
individual needs.
FAMILY CIRCUMSTANCES
 As we have indicated in this and previous chapters, quality of caregiving can
be fully understood only in terms of the larger context of the parent—child
relationship. Job loss, a failing marriage, financial difficulties, and other
stressors can undermine attachment by interfering with the sensitivity of
parental care. Or these stressors can affect babies' sense of security directly,
by exposing them to angry adult interactions or unfavorable child-care
arrangements (Thompson & Raikes, 2003).
 The arrival of a new sibling illustrates how family circumstances can affect
attachment quality.
 In one study, first-born preschoolers who declined in attachment security
after the birth of a baby had mothers who were depressed, anxious, or hostile
before the birth. These symptoms were associated with marital friction
(which the first-borns probably sensed) as well as with unfavorable
mother—first-born interaction. When mothers had cooperative marriages,
coped well with the second birth, and stayed involved with their older child,
preschoolers maintained a secure attachment bond (Teti et al., 1996).
 The availability of social supports, especially a good mother—father
relationship and mutual assistance with caregiving, reduces family stress and
predicts greater attachment security (Owen & Cox, 1997).

PARENTS' INTERNAL WORKING MODELS

 Parents bring to the family context a long history of attachment experiences,
out of which they construct internal working models that they apply to the
bonds established with their babies.
 To assess parents' "state of mind' with respect to attachment, Mary Main and
her colleagues devised the Adult Attachment Interview, which asks adults to
recall and evaluate childhood attachment experiences (Main & Goldwyn,
1998).
 As Table 10.3 shows, quality of mothers' working models is clearly related
to their children's attachment security in infancy and early childhood—
 results replicated in Canada, Germany, Great Britain, the Netherlands, and
the United States.

 Parents who show objectivity and balance in discussing their childhoods

tend to have securely attached infants.
 In contrast, parents who dismiss the importance of early relationships or
describe them in angry, confused ways usually have insecurely attached
babies (Steele, Steele, & Fonagy, 1996; van IJzendoorn, 1995).
 Caregiving behavior helps explain these associations. Mothers with
autonomous/secure representations are warmer and more sensitive with their
babies. They are also more likely to be supportive and to encourage learning
and mastery in their older children, who, in turn, are more affectionate and
comfortably interactive with them (Pederson et al., 1998; Slade et al., 1999).
 But we must be careful not to assume any direct transfer of parents'
childhood experiences to quality of attachment with their own children.
Internal working models are reconstructed memories affected by many
factors, including relationship experiences over the life course, personality,
and current life satisfaction.
 According to longitudinal research, certain negative life events can weaken
the link between an individual's own attachment security in infancy and a
secure internal working model in adulthood. And insecurely attached babies
who become adults with insecure internal working models often have lives
that, based on adulthood self-reports, are fraught with family crises (Waters
et al., 2000; Weinfield, Sroufe, & Egeland, 2000).
 In sum, our early rearing experiences do not destine us to become sensitive
or insensitive parents. Rather, the way we view our childhoods—our ability
 to come to terms with negative events, to integrate new information into our
working models, and to look back on our own parents in an understanding,
forgiving way—is much more influential in how we rear our children than
the actual history of care we received (Main, 2000).
Multiple Attachments: The Father's Special Role
 We have already indicated that babies develop attachments to a variety of
familiar people—not just mothers but also fathers, siblings, grandparents,
and professional caregivers.
 The quality of these attachments can vary, depending on infants' experiences
with each person. Although Bowlby (1969) made room for multiple
attachments in his theory, he believed that infants are predisposed to direct
their attachment behaviors to a single special person, especially when they
are distressed. Consistent with this view, when anxious or unhappy, most
babies prefer to be comforted by their mother. But this preference declines
over the second year. And when babies are not distressed, they approach,
ask to be held by, vocalize to, and smile at both parents equally (Lamb,
1997).
 Like that of mothers, fathers' sensitive caregiving predicts secure
attachment—an effect that becomes stronger the more time they spend with
their babies (van IJzendoorn & De Wolff, 1997).
 And fathers of l to 5 year-olds enrolled in full-time child care report feeling
just as much anxiety as mothers about separating from their child and just as
much concern about the impact of these daily separations on the child's well-
being (Deater-Deckard et al., 1994).
 But as infancy progresses, mothers and fathers in many cultures—Australia,
Indiå, Israel, Italy, Japan, and the United States—relate to babies in different
ways. Mothers devote more time to physical care and expressing affection.
Fathers spend more time in playful interaction—a vital context in which
they build secure attachments with their babies (Lamb, 1987; Roopnarine et
al., 1990).
 Mothers and fathers also play differently. Mothers more often provide toys,
talk to infants, and engage in conventional games, such as pat-a-cake and
peekaboo. In contrast, fathers tend to engage in more exciting, highly
physical bouncing and lifting games that provide bursts of stimulation,
especially with their infant sons (Yogman, 1981).
 In some families, this picture of "mother as caregiver" and "father as
playmate" has changed in response to women's increased workforce
participation.
 Employed mothers tend to engage in more playful stimulation of their babies
than unemployed mothers, and their husbands are somewhat more involved
in caregiving (Cox et al., 1992).
 When fathers are the primary caregivers, they retain their arousing play style
(Lamb & Oppenheim, 1989). Such highly involved fathers are less gender
stereotyped in their beliefs, have sympathetic, friendly personalities, often
had fathers who were more involved in rearing them, and regard parenthood
as an especially enriching experience (Cabrera et al., 2000; Levy-Shiff &
Israelashvili, 1988).
 Fathers' involvement with babies takes place within a complex system of
family attitudes and relationships. Pregnancies were intended rather than
accidental and when both parents believe that men can nurture infants,
fathers devote more time to caregiving.
 A warm, gratifying marital bond supports both parents' involvement, but it is
particularly important for fathers (Braungart-Rieker, Courtney, & Garwood,
1999; Frosch, Mangelsdorf, & McHale, 2000). See the Cultural Influences
box on the following page for cross-cultural evidence that documents this
conclusion—and that also shows the powerful role of paternal warmth in
children's development.